Abstract
Purpose :
Cones in diabetic patients are at risk due to metabolic and vascular changes. The distribution of cones in the macula is difficult to assess due to overlying vessel remodeling and exudation. We are developing a model and sampling method to characterize cone density with reduced artifact from local pathology. We tested whether cones in the temporal meridian of diabetic patients are as dense and regularly distributed, as compared with controls.
Methods :
We compared 10 diabetic subjects (29-79 yr, 54.5 +/- 12.7 yr) to our published norms in 36 younger and 10 older controls. All subjects were consented and tested in a manner approved by the Indiana University Institutional Review Board, which adhered to the Declaration of Helsinki. Custom adaptive optics scanning laser ophthalmoscopy was used to image cones and retinal microcirculation. We sampled temporal retina, 0.9 to 7 deg of 4 non-contiguous samples of nearly 100 x 100 microns, selected for best visibility of cones and least pathology. The data were fit with a 2 parameter exponential model: ln(cone density) = a * microns eccentricity + b. Retinas were also graded based on foveal and temporal data from SD-OCT, and OCT-A (Spectralis). Dilated capillaries, small hemorrhages, and microvascular remodeling were common.
Results :
Diabetic cone maps included bright patches, with unusually bright cones. There were scattered low contrast regions, larger than in older controls. Patchiness is not variation in image quality, which would alter the whole sample. The exponential model provided excellent fits for 7 deg for both younger and older controls: RMS error = 0.9371 +/- 0.0387 and 0.959 +/-0.0267, for younger and older, respectively. RMS errors were 0.767-0.984 in diabetic data sets: 4 of 10 diabetics had RMS error < the 95% confidence limit for either younger or older controls. In 7 of 10 diabetics, the slope parameter, a, was < the 95% confidence limit of younger controls (10.45), including 2 age-matched diabetics. Many diabetics were at the bottom of our older age range, yet had values of a slightly > the 95% cutoff for older controls (10.26). In 2 of 10 diabetics, the offset value, b, was > the 95% confidence limit for younger (-.00050) or older (-.00058) controls.
Conclusions :
There was more variability and lower cone density in several diabetic subjects, particularly near the fovea.
This is a 2020 ARVO Annual Meeting abstract.